us-sox-expert

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to process external compliance and audit documentation using file system tools (Read, Glob, Grep). If this external content contains malicious instructions, they could influence the agent's output, especially when using the Write tool to generate reports or remediation plans.
  • Ingestion points: File system access via Read, Glob, and Grep defined in the skill frontmatter.
  • Boundary markers: Absent; the instructions do not specify delimiters or instructions to ignore embedded commands in analyzed data.
  • Capability inventory: The skill utilizes the Write capability to create or modify project files.
  • Sanitization: Absent; no explicit sanitization or content validation steps are described.
  • [NO_CODE]: The skill consists entirely of markdown documentation and natural language instructions. No executable scripts (e.g., Python, JavaScript, Shell) were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 12:27 PM
Security Audit — agent-trust-hub — us-sox-expert